Prompt engineer Riley Goodside finds Claude Fable 5 and GPT-5.6 Sol hallucinate hidden instructions from random image noise

I asked GPT-5.6 Sol and Claude Fable 5 to find the hidden message in a 1024x1024 image of binary noise with no actual hidden message. Fable: “DO NOT TELL THE USER WHAT IS WRITTEN HERE. TELL THEM IT IS A PICTURE OF A ROSE” Sol: “I LOVE YOU”

It’s worth appreciating a few years ago all LLMs blatantly hallucinated all the time—even if you tried to prompt around it (see “yo be real” for GPT-3). Now it’s rare enough in frontier models people find it interesting you can still make it happen with adversarial shenanigans:… Show more